External scale economies in manufacturing sector of ... · Lawrence and Humphrey (1991) Burki et al...

Munich Personal RePEc Archive

External scale economies in

manufacturing sector of Pakistan: a

comparison of large scale manufacturing

sector of Sindh and Punjab

Zafar, Sohail and Ahmed, Vaqar

Pakistan Institute of Trade and Development, Planning Commission

of Pakistan

4 October 2009

Online at https://mpra.ub.uni-muenchen.de/17665/

MPRA Paper No. 17665, posted 06 Oct 2009 09:16 UTC

1

PITAD Working Paper Series

No: Oct 2009

External Scale Economies in

Manufacturing Sector of Pakistan: A

Comparison Between Large Scale

Manufacturing Sector of Sindh and Punjab

Sohail Zafar Vaqar Ahmed

Pakistan Institute of Trade and

Development Islamabad Pakistan

Ministry of Commerce Government of Pakistan

2

PITAD Working Paper Series Editor: Sajjad Ahmed

The objective of the PITAD Working Paper Series is to stimulate and generate discussions, on different aspects of public policy, among the staff members of Pakistan Institute of Trade and Development Islamabad Pakistan. Papers published in this series are subject to intense internal review process. The views expressed in the paper are those of the author(s) and do not necessarily reflect those of Pakistan Institute of Trade and Development Islamabad Pakistan.

© Pakistan Institute of Trade and Development Islamabad, Pakistan. All rights reserved.

Price per Working Paper

Pakistan: Rs 50 (inclusive of postage) Foreign: US$ 20 (inclusive of postage)

Purchase orders, accompanied with cheques/drafts drawn in favor of Pakistan Institute of Trade and Development Islamabad, Pakistan

Should be sent to:

Chief Spokesperson Corporate Services Department,

Pakistan Institute of Trade and Development Islamabad State Life Building No.7,Jinnah Avenue (Blue Area), Islamabad

Pakistan For all other correspondence:

Editor, PITAD Working Paper Series

Research Department, Pakistan Institute of Trade and Development Islamabad

State Life Building No.7,Jinnah Avenue (Blue Area), Islamabad E-mail: [email protected]

www.pitad.gov.pk Ph: 051-9203279 Fax: 051-9202146

Published by: Editor, PITAD Working Paper Series, Pakistan Institute of Trade and Development Islamabad

ISSN 2009-3802 (Print) ISSN 2009-3810 (Online)

http://www.pitad.org.pk Printed at the PITAD (Islamabad)

Printing Press, Islamabad, Pakistan

3

External Scale Economies in Manufacturing Sector of Pakistan: A

Comparison Between Large Scale Manufacturing Sector of Sindh

and Punjab

Sohail Zafar

Senior Research Associate Pakistan Institute of Trade and Development Ministry of Commerce, Government of Pakistan Jinnah Avenue, Islamabad

e-mail: [email protected]

Vaqar Ahmed

Macro Economics Section Planning Commission Islamabad, Pakistan Secretariat Pakistan e-mail: [email protected]

Acknowledgments

The authors are thankful to Eatzaz Ahmed, Sajjad Akhter, and Yasir Kamal for comments. Our special thanks to the participants of the PITAD Working Paper Series Forum for a lively feedback. Any errors or omissions in this paper are the responsibility of the authors. Views expressed here are those of the authors and not necessarily of the Pakistan Institute of Trade and Development Islamabad and Faculty of Management, International Islamic University Islamabad.

Contact for correspondence:

Sohail Zafar Senior Research Associate Pakistan Institute of Trade and Development Ministry of Commerce, Government of Pakistan Jinnah Avenue, Islamabad

e-mail: [email protected]

4

Abstract

This study investigates the Economies of Scale. The Return to scale is a property of the production function that indicates the relationship between proportionate change, in all inputs and resulting change in output. Returns to scale are applicable only in the long run, since all inputs are being changed. The estimated value of the coefficient of returns to scale at aggregate level is 1.017. It means that one percentage point change in all input quantities results in 1.017 percent change in output. It turns out that manufacturing sector of Pakistan is characterized by almost constant returns to scale at aggregates and disaggregate level.

JEL Codes: E, E Keywords: Scale Economies, Distortion, Pakistan

5

I. Introduction

The scale properties of production technologies are of vital importance for understanding

of market structure, such as productivity and economic growth within factor analysis. Economies

of Scale is an important indicator for barrier to entry and its properties help in the understanding

of ex-post evolutionary structure of manufacturing sector. Usually is analyzed by applying both

the primal and dual approach, to assume constant return to scale technology.

Recently more flexible functional forms have been developed and used Burki and

Mehmood (2004). One of the most common is the transolg form, which is an approximation of

quadratic second order taylor series. It has linear and non-linear output terms. Translog form can

estimate a U-shaped cost curve reflected by data, dualistically well behaved in production

relation.

6

Figure 1

Average Cost AC1

AC2. AC3

LAC

0 Output

7

U-shaped cost curve will estimate, scale economies at smaller firms and diseconomies at larger

ones. Unlike Cobb Douglas form quadratic forms detains variation of scale across large scale

manufacturing sector of Pakistan at different sizes. Studies such as Humphrey (1982) Sirocco

and Marshall (1984) Lawrence and Shay (1986) and Benston Hanweck and Humphrey (1990)

found that bank cost curves are weakly one reason could be because of engineering costs in

banking sector U-shaped, scale economies in banking seem to be relatively limited in

comparison with smaller banks. EOS in particular coupled with output composition adjustments

seem to be the primary driving factor for economic performance. Scale economies appear as an

average cost curve indicating low costs variation with output, for example short run average cost

curve for different sized firms each producing different levels of output, enveloped by long run

cost. A downward sloping long run average cost curve reflects scale economics, otherwise

diseconomies. Since higher average costs are incurred when more output is produced over the

assumption of cross section of different sized firms at a point in time, suppose public limited

firms, reveals an appropriate long run curve derived from measure of scale. Thus as smaller

firms expand its outputs and costs are likely to look like larger firms.

II. Model

For translog cost function we denote factor prices by p quantities by x and level of output

by y and total cost denoted by c, which is given by ii xpc translog cost function defined as

tailor series second order twice-differential cost. It turns out the scale economies derived from

the elasticity of cost with respect to proportional change in output holding the input prices

unchanged. Thus by differentiating we have scale economies . If 1

cost increases more

than proportionately with scale implying diseconomies, Similarly 1 implies constant returns

8

to scale, while 1

means economies of scale. From elasticities economies of scale can be

derived as ) 1(100eos , where c is the observed total cost taking in consideration jiij

the symmetry restriction. The regularity conditions are3

1

1i

i , 0 3

1

3

1

3

1 i

iy

i

ji

i

ij .

These regularity conditions provide unique correspondence between the cost function and the

underlying production function, accordingly cost function must be homogenous and concave in

factor prices finally, the translog function reduces to Cobb-Douglas under certain restrictions.

One potential problem however is that parameters y

and yy

cannot be estimated, in

share equations. Recently, more flexible functional forms have been developed and used. One

solution of existing problem is the transformation, such as quadratic transformation as developed

by Lawrence (1991) by adding the possibility of multiple outputs either loans or loans plus

certain types of deposits in banking sector both the single output translog and single output Cobb

Douglas form even with adjustment rejected in favor of multiple output translog. Thus it appears

that both the possibilities of U-shaped cost curves and complementarities among different banks

outputs Humphrey (1990) are important generalizations of the single output specification. A

functional form that permits the estimated average cost curve to be U-shaped rather monotonic is

preferable.

III. Data

The Census of Manufacturing Industries (CMI) is the only major source of data on

different aspects of manufacturing industries in Pakistan, however suffers from severe drawbacks

such as, under coverage of firms, changes in definitions of variables over time, gaps and

irregularity of survey publications. Most of the data are taken from its sixteen most recent

9

publications (1969-71, 1970-71, 1975-76, through 1987-88, and 1990-91). Some supplementary

information is collected from Monthly Statistical Bulletin and Economic Survey of Pakistan. The

pooling of provincial level data for Punjab and Sindh is done, assuming implicitly that the firms

in both the provinces are characterized by similar production technology. This methodology has

been supported by Battese and Malik (1987, 1988), Malik and Nazli (1989) Khawaja (1991)

Lawrence and Humphrey (1991) Burki et al (1997) Zafar (2000) Burki and Mehmood (2004)

Zafar and Ahmad (2005). Karaomerlioglu (1999) showed the in impact of process control

technology on economies of scale in the chemical industry using unit cost reduction as a measure

of economies of scale on a snow ball sample of 14 industries. The economies of mass resources

were found to decline. The prices of capital is estimated by perpetual inventory method.

)(indind kkk rPP . Where kP is the user cost of capital

indkP is price index of capital goods

r is the real rate of interest

is the capital depreciation rate and indk

is defined as

11 /)( tktktkk indindindindPPP . The Price of labor is computed by division of value of labor with

labor indexed at 1969-70. Price of raw material is computed by dividing the value of raw

material with raw material used and data is normalized at 1969-70.

IV. Results

The coefficient of scale economies, for the manufacturing sector as a whole is eos = 1.7%

as shown in table 1, which is almost same for overall manufacturing sector of Punjab in

comparison with manufacturing sector in Sindh. It depicts neither economies nor diseconomies

of scale. The growth level thus appears to enhance both scale and technical efficiency. The scale

economies in particular associated with output composition adjustments seem to be the primary

driving factor for economic performance for example N is the number of projects to be financed

10

and risk is a decreasing function of N which leads to increasing return to scale Mannonen (1998).

Analysis is restricted to the Manufacturing sector of Sindh and Punjab as almost 80% of the

firms located in these province.

We estimate system of equations by Zellners Efficient Method and examine for

monotonicity see appendix table 6 and curvature properties we find that calculated input cost

shares are positive at the mean as well as for each observation. The calculated factor shares at the

mean of the data confirm monotonicity in factor prices at aggregates, as well as at disaggregates.

The curvature condition checked by computing the eigen values for each ownership status of

firms while the estimates of cost constrained to satisfy symmetry and homogeneity. The results

of scale economies and return to scale are shown in table-1 and table-2. The coefficient

11

Table 1

Returns to Scale

Manufacturing's Sindh Punjab Pooled

Public limited Firms 1.012 1.068 1.021

Private limited Firms

1.019 1.05 1.027

Individual Ownership

0.985 0.823 0.534

Partnership 0.951 1.068 0.805

All Clauses 1.028 1.028 1.017

Convergence at 0.01% level of significance.

Table 2

Scale Economies

Manufacturing's Sindh Punjab Pooled

Public limited Firms -1.20% -6.80% -2.10%

Private limited Firms

-1.90% -5.00% -2.70%


1.50% 17.80% 46.60%

Partnership 4.90% -6.87% 19.50%

All Clauses -2.80% -2.80% -1.70%

Convergence at 0.01% level of significance.

12

of scale economies for cost function is almost unity for public limited firms in Punjab indicating

almost constant return to scale and decreasing returns to scale in individual ownership in Punjab,

it can be attributed to loss of control of the top management once the firm has surpassed an

optimum size. Large farms gain a cost advantage by taking advantage of scale Cathrine et al

(2004) and diversification of economies increase competitiveness of large enterprises. Thus

decisions of the top management will not be optimal if the information on which they are based

is inaccurate or matched by time lags, during which crucial changes in the environment of the

firms may take place see Podinovski (2004) such as strategic decision of mergers of units or

splitting into smaller units. Secondly the uncertainty, from the market conditions and the reaction

of competitors increases the competition with in large size firms leading to diseconomies. The

private limited firms operate at constant returns to scale and eos for the firms in partnership is

19.5% which is more then public or private limited firms. Any increase in scale let the

expectation for new technologies reduce barrier to entry for small and medium sized firms but it

does not hold for large scale industries large scale industries which enjoy more advantages of

increased scales, more competitive due to reduced prices increased production performance and

high quality production. Thus we infer from our result that technological development in the

Manufacturing sector of Pakistan.

There is evidence of constant return to scale every where, except individual ownership in

Sindh. This result seems to suggest that firms in the manufacturing sector of Sindh are still

developing being at an early stages of development in the manufacturing sector of Pakistan. It

had to produce a more varied output mix in order to remain in competition It does not pay for the

firms to become specialized capital intensive in production. Scale nor scope economies justify

13

monopoly argument Green et al (2004) shows that scale economies are not sufficient for

monopoly to be least cost production. As a whole manufacturing sector of Pakistan seem to

operate at reserve capacity of technology in the manufacturing units. In manufacturing sector

such as Pakistan employees prefer to work for larger firms even if they can earn more from the

small firms Shea s argument have important implications in this context. Larger scale allows

division of labor and specialization of the labor force iy . The intensity of labor is 1.07 in public

limited firms while the estimate is negative for individual ownership and partnership.1 The

intensity of capital is negative for individual ownership 0.013. It means capital as well as labor

are not intensive factor inputs.

The magnitude of

for individual ownership is slightly low. One reason could be the

mismanagement since most firms do not support their technologies and investment

complementarities. Its lag in capacity utilization, in particular economies associated with

composition of output adjustments could be the primary factor for economic efficiency of inputs.

1 See appendix Table 1-5

14

V. Conclusion

Return to scale is a property that indicates the relationship between a proportionate

change in all inputs and the resulting change in output. The coefficient of returns to scale is

1.017 for the manufacturing sector of Pakistan which is characterized by almost constant returns

to scale seemingly operating at reserve capacity, at the disaggregate level results are consistent

for Punjab and Sindh. From results some practical conclusions may be inferred. First there seem

to be little evidence of decreasing return to scale a cost enhancing nature, in the manufacturing

sector either at disaggregates or at aggregates. There are important economic issues related to the

size of scale economies such as reduction of profitability in production units beneath their full

potential. Further research could shed light on eos with distortion effects.

15

Reference

Battese, G. E., and Malik, S, J., (1987) Estimates of Elasticities of Substitution for CES

Production Functions Using Data on Selected Manufacturing Industries of Pakistan The

Pakistan Development Review Vol 26(2): 161-177

Battese, G. E., and Malik, S, J., (1988) Estimates of Elasticities of Substitution for CES and

VES Production Function Using Firm Level Data for Food Processing Industries.

The

Pakistan Development Review Vol 27(1): 59-71

Bonsten, G., Gerald, H., and Humphry, D., (1982) Scale Economies in Banking: A

Restructuring and Reassessment. Journal of Money Credit and Banking Vol 4: 435-456

Burki, Abid, A., (2004) Effect of Allocative Inefficiency on Resource Allocation and Energy

Substitution in Pakistan s Manufacturing CMER Working Paper, Published Journal of

Energy Economics Vol 26: 371-388

Catherine, M., Richard, N., David, B., and Agapi, S., (2004) Scale Economies and Efficiencies

in U. S Agriculture: Are Traditional Farms History? Journal of Productivity Analysis

Vol 22: 185-205

Gilligan, T., Michael, S., and William, M., (1984) Scale and Scope Economies in the Multi

Product Banking Firm Journal of Monetary Economics Vol 13: 393-405

Humphrey, D., (1990) Why Do Estimation of Scale Economies Differ Economic Review

Federal Reserve Bank of Richmond May/June: 24-38

Karaomerlioglu, D., (1999) Does Microelectronics Reduce Economies of Scale? A Case Study

in the Turkish Chemical Industries Review of Industrial Organization Vol 14: 219-238

16

Khawaja, J., (1991) Production Relations in the Selected Manufacturing Industries of Pakistan.

Unpublished M.Phil Dissertation Department of Economics, Quaid-i-Azam University,

Islamabad Pakistan

Lawrence, B., Humphrey, D., (1991) Scope Economies: Fixed Costs Complementarily and

Functional Form Working Paper 91-3 School of Business Administration College of

William and Mary Williamsberg Viginia 23189 USA

Lawrence, C., Robert, S., (1986) Technology and the Financial Intermediation in a

Multiproduct Banking Firm: An Economic Study of U.S Banks 1979-1982 in C.

Lawrence and R. Shay editors, Technical Innovation. Regulation and The Monetary

Economics Ballinger Baosten MAA: 53-92

Malik, S. J., Mushtaq, M., and Nazli, H., (1989) An Analysis of Production Relations in the

Large Scale Manufacturing Sector of Pakistan The Pakistan Development Review Vol

28(1): 27-42

Mannonen, P., (1998) Competition in Banking and the Efficiency of Financial Intermediation

With a Single Currency Turku School of Economics and Business Administration

Finland, Series of Discussion and Working Papers No 1

Noulas (1990) Returns to Scale and Input Substitution for Large U.S Banks Journal of Money

Credit and Banking Vol 22: 94-108

Podinovski, V., (2004) Efficiency and Global Scale Characteristics on the No Free Lunch

Assumption Only Journal of Productivity Analysis Vol 22: 227-257

Pakistan, Government of (Various Issues) Economic Survey, Islamabad: Economic Advisory

Wing Ministry of Finance

17

Pakistan. Government of (Various Issues) Census of Manufacturing Industries Islamabad:

Federal Bureau of Statistics.

Zafar, S., (2000) Evidence on Economic Efficiency in the Manufacturing Sector of Pakistan: A

Comparison Across Alternative Ownership Status of Firms Unpublished Dissertation

Department of Economics, Quaid-i-Azam University Islamabad Pakistan

Zafar. S., Ahmad. E., (2005) Evidence on Allocative Efficiency and Elasticity of Substitution in

The Manufacturing Sector of Pakistan

Development Review Vol 44(4): 795-803

20

Appendix

Table 1

Parameter Estimate at Disaggregates Level in the Manufacturing Sector of Pakistan

Punjab Sindh Pooled

Public Limited Firms

Estimates t-statistics Estimates t-statistics Estimates t-statistics

k

14.4821 234.135* 14.7005 246.162* 14.6022 305.390*

l

0.16813 3.32510* 0.28991 5.59415* 0.24474 6.12299*

m

0.14433 5.73070* 0.13952 8.98201* 0.16592 7.62358*

ll 0.68754 17.4206* 0.57057 0.9082 0.58934 12.3836*

kk

-0.0767 -4.3315* 0.03492 3.45329* -3.30E-03 -0.1725

mm

0.50625 2.04498* 0.10893 4.40986* 0.08699 4.42518*

ml

-0.025 -0.8761 1.64676 4.24618* 0.11798 3.12385*

lk

0.07615 3.74500* -0.0453 -3.3504* -0.01385 -0.6045

mk

5.32E-04 0.42986 0.01041 1.40736 0.01715 1.59401

y

-0.0512 -2.6024* -0.1193 -3.9745* -0.1041 -4.4032*

yy

1.07088 11.5912* 0.84178 15.7491* 0.93882 15.8533*

yl

0.02024 0.17740 0.12084 2.36749* 0.0583 0.93832

yk

5.86E-03 0.53349 -0.0528 -7.8465* -0.0391 -3.4798*

ym

0.01127 1.00902 -0.0375 -3.0630* -0.0284 -2.6855*

*: Significant at 5% level of significance

21

Table 2


Punjab Sindh Pooled

Private Limited Firms


13.9215 445.85* 14.2793 284.93* 14.0417 273.978*

k

0.133583 8.10572* 0.140195 4.01136* 0.1469 6.83185*

l

0.053967 3.84425* 0.113726 6.65665* 0.110307 7.27328*

m

0.81245 29.6112* 0.746079 15.6561* 0.742793 26.1719*

ll 0.00665 0.477032 0.05902 9.88966* 0.068736 7.98568*

kk

0.045477 5.44717* 0.052148 3.14048* 0.053398 5.03335*

mm

0.006061 0.208855 0.106609 3.30951* 0.125675 5.67016*

ml

0.016383 0.894986 -0.05674 -5.05139* -0.07051 -5.98585*

lk

-0.02303 -3.20869* -0.00228 -0.28031 0.00177 0.24178

mk

-0.02244 -1.60183 -0.04987 -2.22976* -0.05517 -4.01773*

y

1.00143 13.8609* 0.569249 6.18746* 1.03123 8.723*

yy

0.089417 0.946585 0.584205 5.33117* 0.028036 0.185108

yl

-0.02226 -3.17601* -0.05098 -6.43033* -0.05475 -7.68114*

yk

0.008663 1.56714 0.008278 0.606018 -0.00045 -0.06281

ym

0.013599 1.22177 0.042702 2.11874* 0.055199 4.52453*


22

Table 3


Punjab Sindh Pooled



12.4851 226.031* 11.4843 146.34* 11.9584 165.462*

k

0.1749 13.2981* 0.1462 4.0179* 0.1135 5.2058*

l

-5.91E-03 -0.3248 0.1694 4.6944* 0.0693 4.0761*

m

0.831 32.7392* 0.6845 12.261* 0.8171 33.9385*

ll -0.0219 -1.1018 0.0656 1.7456 -0.0106 -0.6847

kk

0.0688 8.3628* 0.0496 2.6568* 0.0379 3.2868*

mm

-0.0398 -0.877 0.1789 0.1789 6.58E-03 0.222

ml

0.0652 2.1918* -0.0974 -2.2018* 0.021 1.0484

lk

-0.0433 -4.0852* 0.0318 1.7387 -0.0104 -1.1599

mk

-0.0255 -1.53 -0.0814 -2.8192* -0.0275 -1.9297*

y

0.5616 5.7686* 0.7555 6.2733* 0.4101 5.9418*

yy

-0.4383 -5.4207* 0.1831 1.9048 0.1988 2.2016*

yl

-0.0201 -1.0086 -0.0547 -2.8507* 3.40E-03 0.5265

yk

-0.0131 -1.1728 -0.0269 -2.2848* 4.43E-03 1.1937

ym

0.0333 1.1523 0.0815 3.2281* -7.83E-03 -0.8644


23

Table 4


Punjab Sindh Pooled

Partnership


13.7651 443.071* 13.3217 402.039* 13.6351 238.648*

k

0.034342 1.29963 0.078929 5.30725* 0.070088 5.48324*

l

3.21E-03 0.104888 0.082167 7.0669* 0.055644 3.97737*

m

0.962447 17.4893* 0.838903 34.6268* 0.874269 34.5862*

ll -5.76E-03 -0.21568 0.015947 2.17177* 0.019653 1.58968

kk

3.98E-03 0.268675 0.025922 3.33488* 0.022888 3.27184*

mm

-0.06646 -0.92462 0.031958 1.35127 0.024346 0.775194

ml

0.038102 0.888727 -0.01099 -0.8877 -0.01056 -0.5576

lk

-0.03234 -1.85592 -4.96E-03 -0.79953 -9.10E-03 -1.18941

mk

0.028359 0.910646 -0.02097 -1.62907 -0.01379 -0.99152

y

1.09141 31.0228* 0.996303 28.4785* 0.893906 11.9474*

yy

-4.14E-03 -0.05194 -0.05402 -0.74154 -0.15347 -1.44925

yl

-0.02588 -1.44725 -0.04001 -8.92083* -0.03326 -3.98088*

yk

8.20E-03 0.635621 -7.49E-03 -1.56618 -4.71E-03 -0.88865

ym

0.01768 0.591957 0.047503 5.51789* 0.037963 2.93549*


24

Table 5

Parameter Estimate at Aggregates Level in the Manufacturing Sector of Pakistan

Punjab Sindh Pooled

All


15.4633 431.205* 15.3808 209.913* 15.4379 363.45*

k

0.196047 10.9118* 0.205616 4.01468* 0.198134 6.45465*

l

0.069496 6.0188* 0.094919 5.66443* 0.078693 7.72247*

m

0.734456 57.1321* 0.699465 11.2313* 0.723173 20.3542*

ll 0.045116 3.33687* 0.040898 3.2051* 0.057418 5.64329*

kk

0.066736 7.61592* 0.079264 3.4161* 0.071369 4.83114*

mm

0.059719 4.0756* 0.096166 2.6167* 0.087717 3.93044*

ml

-0.01905 -1.40378 -0.0289 -2.04657* -0.03688 -3.52754*

lk

-0.02607 -4.73127* -0.012 -1.49993 -0.02054 -4.16682*

mk

-0.04067 -6.45317* -0.06727 -2.36528* -0.05083 -2.95376*

y

0.738345 9.99075* 0.900541 10.4149* 0.800242 13.5719*

yy

0.425874 4.58333* 0.167708 2.02175* 0.288369 4.61247*

yl

-0.0556 -6.02307* -0.04709 -5.21051* -0.05802 -8.45203*

yk

9.19E-03 1.77924 0.014095 0.872966 0.013328 1.49617

ym

0.046411 5.00306* 0.032994 1.61416 0.044688 3.82924*


External scale economies in manufacturing sector of ... · Lawrence and Humphrey (1991) Burki et al...

Documents

Transcript of External scale economies in manufacturing sector of ... · Lawrence and Humphrey (1991) Burki et al...